Recent years have seen a significant technical development in digital video apparatuses and demands for compression-coding a video signal (a plurality of pictures arranged in time series). Such a compression-coded video signal is, for example, recorded on a recording medium such as a DVD and a hard disc, and is distributed on a network. The H.264/AVC (MPEG-4 AVC) is one of the image coding standards, and, as the next-generation standard, the High Efficiency Video coding (HEVC) standard is currently being considered (Non-patent Literature 1).
The HEVC standard today involves a step of predicting an image to be coded, a step of calculating a residual between a coding-target image and a prediction image, a step of transforming the residual image into frequency coefficients, and a step of performing arithmetic coding on the frequency coefficients. In the arithmetic coding step, context adaptive arithmetic coding is performed on the components (coefficients) included in the coding-target block in the order from highest frequency components to low frequency components.